86 CANADIAN JOURNAL O F CHEMISTRY. VOL. 46, 1968

A rapid, mild procedure for the preparation of alkyl chlorides and bromides

J. Hooz AND S. S. H. GILANI Depflrtment of Cl~elnistr.~, Ut~iversity of Alberta, Erlnlontotr, Alberta

Received May 17, 1967

Primary and secondary alkyl chlorides have been conveniently prepared by the reaction of tri-n- octylphosphine with carbon tetrachloride solutions of the corresponding alcohols. This rapid, high yield reaction proceeds with inversion of configuration. By using carbon tetrabromide the method has been extended to the synthesis of alkyl bromides.

Canadian Journal of Chemistry, 46, 86 (1968)

The formation of ylid 1 (R = C6H5) and trivhenvlvhosvhine dihalide (2) from the inter- [41 R~$OR' 2 -> R'X + R3P0

aciion oi trip~ienylphosphine with carbon tetra- Accordingly, we have found that phosphines halides (eq. [l]) was simultaneously and inde- snloothly convert alcohols to chlorides (eq. [5]). pendently discovered by Rabinowitz and Marcus (1) and Ramirez and co-workers (2). An ionic

mechanism iilvolving nucleophilic displacement on halogen has been invoked (3) to account for the formation of these products (eqs. [2] and [3]).

Ylids of structure 1 (R = C6Hj) possess considerable syntlietic potential since they provide a route to otherwise difficultly accessible 1,l-dihaloalkenes on reaction with carbonyl compounds (1,2,4). Reagents of the type R3PX2 (R = C6H5 or 12-C4H9) have been demonstrated to perinit the conversion of alcol~ols to alkyl halides to proceed generally without complica- tion of elii1liilation or rearrangement (5, 6); in addition, they convert phenols to aryl ha!ides (6,7) without formation of position isomers.

The present study arose out of a consideration that intermediates of structure 3 (or 4) C O L I ~ ~ be trapped by reaction with alcohols to provide 5 (or 6) , e.g.

The anticipated collapse of 6 to alkyl halide and tertiary phosphine oxide finds ample analogy in a related study (8).

[5] R3P + R'OH + CCI4 + R'C1 + R3P0 + HCC13 where R = t1-C8Hl7 or C6H5

Using tri-12-octylphosphine (TOP) and primary al- cohols, conversions are of the order 90-100 % in favorable cases. Tripl~enylphospl~ine may also be used at longer reaction times.1 Although no at- tempts at optin~ization have been made, we find that this facile transformation is conveniently ac- complished (for chlorides) by adding a slight ex- cess over one inole equivalent of TOP to a solution of alcohol in carbon tetracl~loride as solvent. A vigorous reaction ensues which is con~plete, in the case of primary alcohols, in ca. 5 min. The product is readily isolated by standard methods of distillation and chromatograpl~y (cf. Experi- mental).

Typical results are the formation of i2-C5H1 lC1 (94%), C6H5CH2CI (loo%), rz-C8HI7Cl (93 %), and C6H5CH2CH2C1 (66 %).2

Secondary alcohols also react. Thus, after ca. 5 mill reaction time, sec-butyl chloride and 2-chlo- rooctane were formed in 60 % and 80 % yields, respectively, from the corresponding alcohols. When (+)-2-octanol was subjected to this reaction, (-)-2-chloroocta~~e was obtained in 90% optical purity, a result consistent with a bimolecular displacement process acco~npanied by a Walden inversion in the product determin- ing step (eq. [4]).

The incursion of a competing olefin-forming elimination process renders this method of

IRecently, and concurrent with the present work, a similar study has been reported (10) using triphenyl- phosphine.

2Analysis by gas-liquid partition chromatography.

Can

. J. C

hem

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

JOU

RN

AL

IN

FOR

MA

TIO

N S

EC

TIO

N o

n 02

/07/

13Fo

r pe

rson

al u

se o

nly.

limited value for the preparation of tertiary chlorides. For example, t-butanol gave only 13-15% t-butyl chloride, and isobutylene was detected by gas-liquid partition chromatography (g.1.p.c.).

Bromides may also be prepared by this method but in this case the ratio of reagents is ROH: TOP:CBr4 = 1 :2:2. Diethyl ether is a suitable solvent. In this manner n-C5H11Br (97%) and C6H5CH2Br (96%)2 were formed after ca. 5 min reaction time.

In view of the mild conditions employed and good yields obtained, this method offers con- siderable promise as a synthetic tool.

Experimental All melting and boiling points are uncorrected. The

alcohols, chlorides, and bromides were commercially available materials whose homogeneity was established by g.l.p.c., using an Aerograph A90-P3 instrument equipped with a thermal conductivity detector. Infrared spectra were recorded on a Perkin-Elmer model 421 spectrophotometer. Nuclear magnetic resonance spectra were recorded on a Varian A-60 analytical spectrometer using tetramethylsilane as internal standard.

All conversions of alcohol to akyl chloride were accomplished by a procedure similar to that employed for the preparation of 1-chlorooctane which is described in detail. Bromides were prepared by the method des- cribed for the synthesis of 1-bromopentane.

Preparation of1 -Ci~lorooctat~e To a magnetically-stirred solution of 1-octanol (10.0

mmoles) in 20 ml carbon tetrachloride was added tri-n- octylphosphine (10.5 mmoles). An exothermic reaction ensued. (In larger scale runs it is recommended that the reaction be modcrated by a cooling bath.) Approximately 5 min after complete addition of the phosphine an aliquot was analyzed by g.1.p.c. (5 ft x 1 /4 in. FFAP on Chromo- sorb W column operated at 140°, helium flow 55 ml/min). No octanol was present, but a new peak appeared, and was shown to be I-chlorooctane by comparison of retention timc with that of an authentic sample. The yield3 (as determined by g.1.p.c.) was 94%. By operating at a column temperature of 50" and helium flow of 25 ml/min, the peaks due to carbon tetrachloride and chloroform could be resolved but the yield of the latter was' not determined.

Chloroform and carbon tetrachloride were distilled through a 25 cm Vigreaux column and the residue was chromatographed on BDH alumina. The pentane eluate

3Under otherwise identical conditions, the yield of 1-chlorooctane was 47 % after 5 min reaction time cmploy- ing trlphenylphosphine.

was concentrated and distilled to give an 88% yield of pure 1-chlorooctane.4

Zsolatiotz of Tri-tz-octylphospizit~e Oxide The above-described experiment was duplicated, and

after ca. 5 min reaction time the crude mixture was distilled to remove all material boiling up to ca. 184'. The residue was triturated with pentane and cooled to -78". The resulting solid was filtered, washed with cold pentane, and dried. In this manner there was obtained 3.55 g (92%) of solid, m.p. 48-52" (lit. (9) m.p. 48-54"), vmns (Nujol) 1138 cm-1 (PO). The nuclear magnetic resonance spectrum (CDC13) showed absorption for 9 protons as a multiplet centered at T 9.1 (CH3) and 42 protons with absorption from ca. T 8.0-8.92 (CH2).

Preparatiot~ of I-Brornopentane Tri-/I-octylphospl~ine (20 rnmoles) was added to a

magnetically-stirred solution of 1-pentanol (10 mmoles) and carbon tetrabromide (20 mmoles) in 20 ml anhydrous ether. An immediate exothermic reaction took place and the initially colorless mixture turned yellow. Fivc minutes after complete addition of the phosphine an aliquot was analyzed by g.1.p.c.; the yield of l-bromo- pentane was 97%. The mixture was worked up in a manner identical to that described for 1-chlorooctane, and upon distillation there was obtained an 80% yield of 1-bromopentane.4 Since no special precautions were taken to minimize handling losses this yield can doubtless be improved.

Ackno%vIedgment The financial assistance of the National

Research Council of Canada is gratefully acknowledged.

I . R. RABINOWITZ and R. MARCUS. J. Am. Chem. Soc. 84,1312 (1962).

2. F. RAMIREZ, N. B. DESAI, and N. MCKELVIE. J. Am. Chem. Soc. 84,1745 (1962).

3. B. MILLER. Topics Phosphorus Chem. 2,133 (1965). 4. A. J. SPEZIALE, G. J. MARCO, and K. W. RATTS.

J. Am. Chem. Soc. 82, 1260 (1960); A. J. SPEZIALE and K. W. RATTS. J. Am. Chem. Soc. 84, 854 (19621.

5. L. HORNER, H. OEDIGER, and H. HOFFMANN. Ann. 626,26 (1959).

6. G. A. WILEY, R. L. HERSHKOWITZ, B. M. REIN, and B. C. CHUNG. J. Am. Chem. Soc. 86,964 (1964).

7. H. HOFFMANN, L. HORNER, H. G. WIPPEL, and D. MICHAEL. Chem. Ber. 95.523 (19621.

8. J. P. SCHAEFER and D. S. WEIN~ERG.' J. Org. Chem. 30,2635 (1965).

9. C. A. BLAKE, K. B. BROWN, and C. F. COLEMAN. U.S. At. Energy Comm. ORNL-1964 (1955); Chem. Abstr. 50, 15321b (1956).

10. I. M. DOWNIE, J. B. HOLMES, and J. B. LEE. Chem. Ind. London, 900 (1966).

"he structures of all synthetic materials were secured by comparison of physical properties with those of authentic samples.

Can

. J. C

hem

. Dow

nloa

ded

from

ww

w.n

rcre

sear

chpr

ess.

com

by

JOU

RN

AL

IN

FOR

MA

TIO

N S

EC

TIO

N o

n 02

/07/

13Fo

r pe

rson

al u

se o

nly.